Rescuing the MDGs: Paying for resultsNYU, New YorkSeptember 15, 2005
Owen Barder and Nancy BirdsallCenter for Global DevelopmentWashington, D.C.
Outline
What’s wrong with the MDGs?
The current approach
Rescuing the MDGs: What about aid?
The proposal
Trans-Atlantic tension: A synthesis
What’s wrong with the MDGs?
o They are not realistic…
What’s wrong with the MDGs?
Source: Clemens and Moss (2005). “What’s Wrong with the Millennium Development Goals” at http://www.cgdev.org/content/publications/detail/3940
What’s wrong with the MDGs?
They are not realisticAnd they risk turning “real” success into “imaginary” failure
What’s wrong with the MDGs?
Source: Clemens and Moss (2005).
The current approach
The current approach
Starts from needs (good)Adds up “costs” (misleading)
The current approachMDG cost estimate for meeting the MDG education goal1
(per capita, constant 2003 US$)
Education2 Primary education3
2006 11 -2010 17 -2015 25 -2006 15 -2010 19 -2015 22 -2006 17 132010 19 152015 22 172006 11 82010 13 102015 17 122006 14 72010 15 82015 17 12
Notes:
Uganda
1. MDG investment needs are defined as the capital investments and operating expenditures required to meet a given MDG, excluding expenditures for capacity building.2. Education estimates include primary and secondary education.3. The primary education cost estimates are in constant 2000 US$ and the 2006 numbers refer to 2005 for this indicator.Source: UN Millennium Project (2005) Investing in Development. A Practical Plan to Achieve the Millennium Development Goals, Chapter 17 and UN Millennium Project (2004) "Millennium Development Goals Needs Assessments for Ghana, Tanzania, and Uganda."
Bangladesh
Cambodia
Ghana
Tanzania
The current approach
But “cost” estimates ignore demand constraints, complementarities, non-marginal, non-constant returns…
And they suggest more spending is the answer
Though there is no particular association between spending and education outcomes
The current approach
Bangladesh
Armenia
Azerbaijan
Belize
Bolivia
Cambodia
Cape Verde
ChileColombia
Costa RicaCroatia
Dominican Republic
El Salvador
Gambia, The
Georgia
Kazakhstan
Kyrgyz Republic
Lao PDR
Lebanon
Macao, China
Mauritius
Moldova Mongolia
Morocco
Namibia
Nicaragua
Oman
Panama
South Africa
Swaziland
Togo Trinidad and Tobago
UkraineUnited Arab Emirates
United Kingdom
United States
7080
9010
0S
choo
l enr
ollm
ent,
prim
ary
(% n
et)
2 4 6 8 10Public spending on education, total (% of GDP)
Source: WDI (2005).
The current approach
. . . not in the developing world, and not in New JerseyAbbott v. Burke caseEqualize school funding: New JerseyEqual since 1997Transfer of $1,924 per student$30 billion transferred, 1997-2004
Source: Clemens (2005) “International Tax.”
New Jersey 8th grade basic language skills test: pass rate
41%47%
74%80%
0%
25%
50%
75%
100%
1996 2004
Abbott districtsRest of state
Source: Clemens (2005) “International Tax.”
The current approach
Camden State
African-American/Latino 97% 33%Avg. residence value $24,400 $147,500Eligible for free lunch 78% 28%Crime rate per 1000 86 32Unemployment 12% 4%Budget per student $17,500 $12,0008th grade proficiency test 25% 80%
Source: Clemens (2005) “International Tax.”
The current approach
“Cost estimates ignore demand constraints, complementarities, non-marginal, non-constant returns…
Suggest more spending is the answer…
And the need for more spending implies foreign aid is the constraint
The current approachAs many studies seem to imply…
Source: Clemens and Moss (2005)
Study Covering Additional resources / year
Global poverty goal $54-62 billion Global social and environ. goals $35-75 billion Devarajan, et al "Goals for Development" (WB
2002) Global primary education $10-15 billion Vandemoortele, "Are the MDGs feasible?" (UN 2002)
Global MDGs $50-80 billion
Global poverty goal $15-46 billion, + cancel debt Greenhill, "The Unbreakable link" (Jubilee
Research, 2002) Global, other goals $16.5 billion, + cancel debt
"Last Chance in Monterrey" (Oxfam 2002) Global MDGs $100 billion Asia and South Asia double / triple aid "Supporting Sound Policies…" (WB 2003) Africa and Central Asia 60% increase
"Achieving the MDGs in Africa…" (AfDB 2002) Sample of 30 African countries $20-25 billion Delamonica et al (Unicef, 2001) Global primary education $9.1 billion
Primary education, 47 IDA countries
$2.5-5 billion "Education for Dynamic Economies…" (WB 2002) African primary education 7x aid Naschold, "Aid and the MDGs" (ODI 2002) Global primary education $9 billion Filmer, “Costing the Goal…” (WB 2002) Global Primary education $30 billion Mingat, et al "Financing Education for all…" (WB 2002)
Primary Education for 33 African countries
$2.1 billion
Brossard et al., “Education Primaire Universelle: Combien?” (Unesco 2001)
African primary education $2.9-3.4 billion
Bruns et al, A Chance for Every Child (WB 2003)
Low-income primary education $5-7 billion
Rescuing the MDGs: What about aid?
Aid cannot drive the process…but constraints on resources should not be a constraint on progressAid should create and enhance incentives for households, communities, and governments toward the MDGs, and support good institutions
Rescuing the MDGs…(and the aid business from itself)
Rescuing the MDGs: What about aid?
But currently aid tends to make recipient countries accountable to donors, not their citizens
Total net ODA
(% of GNI)
Total net ODA per capita
(current US$)
Government revenue excl. grants(% of GDP)
Burundi 39 31 21Sierra Leone 39 56 12Malawi 30 45 18West Bank and Gaza 25 289 -Mozambique 25 55 14Ethiopia 23 22 20Nicaragua 21 152 21Mauritania 21 85 -Rwanda 20 40 14Mongolia 20 100 38Niger 17 39 10Tanzania 16 47 11Uganda 16 38 12Zambia 13 54 18Source: WDI (2005), ADB (2005) and IMF Regional Economic Outlook for Sub-Saharan Africa (2005).
Aid and government revenue in selected countries 2003
The proposal
Recipient government defines country-specific goals (not “universal MDGs)Agrees with donor community on country-specific benchmarks Donors tie aid tied to progress against those agreed benchmarksBenchmarks can be inputs, outputs, outcomes (all constitute “results”)
The proposal
Donors agree on a legally binding tariff, e.g. $50 per year for every child in school (above 1990 enrollment)
Cash amounts the same for all countries
Annual independently audited “invoice” presented by governments to donors
The proposal: An education example
MDG goal: all children complete primary school by 2015
MP estimates of “cost” in poorest countries of putting every child in school
. . . provide a “weak” rationale for a payment of $50 per child
(Back to education “costs”)MDG cost estimate for meeting the MDG education goal1
(per capita, constant 2003 US$)
Education2 Primary education3
2006 11 -2010 17 -2015 25 -2006 15 -2010 19 -2015 22 -2006 17 132010 19 152015 22 172006 11 82010 13 102015 17 122006 14 72010 15 82015 17 12
Notes:
Uganda
1. MDG investment needs are defined as the capital investments and operating expenditures required to meet a given MDG, excluding expenditures for capacity building.2. Education estimates include primary and secondary education.3. The primary education cost estimates are in constant 2000 US$ and the 2006 numbers refer to 2005 for this indicator.Source: UN Millennium Project (2005) Investing in Development. A Practical Plan to Achieve the Millennium Development Goals, Chapter 17 and UN Millennium Project (2004) "Millennium Development Goals Needs Assessments for Ghana, Tanzania, and Uganda."
Bangladesh
Cambodia
Ghana
Tanzania
The proposal
The process should create incentives for households (enhance household demand), support public “supply”, and make government accountable to its own citizens.
Vouchers worth $50 go to poor households for each child in school,
Are collected by schools, which
“sell” them to local government or other official intermediary. . . and then to the Finance Ministry, which bills donors
The proposal
Donor transfers at actual primary enrollment 2001
($50 per student)
Donor transfers for 1990-2001 increment in primary
enrollment
($50 per student)
Donor transfers for 1990-2015 increment in primary
enrollment assuming universal enrollment in 2015($50 per student)
Bangladesh 870,000,000 255,000,000 500,000,000
Cambodia 85,000,000 33,000,000 58,000,000
Ghana 105,000,000 29,000,000 115,000,000
Tanzania 155,000,000 23,000,000 237,000,000
Uganda - - 310,000,000
Tajikistan 50,000,000 20,000,000 17,000,000Sources: United Nations World Population Prospects: The 2004 Revision, World Bank EdStats (2005) and author's calculations.
Primary educationRough estimates of resulting annual donor transfers, selected countries
The proposal
Benefits:
No arguments over “conditionality.” Payments tied to results not policies
Binding donor commitments provide predictable, long-term funding against which performing governments can plan, invest…
and borrow on private markets
Makes recipient country governments accountable since a shortfall in achievement cannot be blamed on insufficient donor aid.
The proposal
“Changing education systems requires political leadership and institutional reform, as well as additional investments and inputs”
Source: UN Millennium Project ,Taskforce on Education and Gender, Toward Universal Primary Education: Investments, Incentives and Institutions, p. 24.
Trans-Atlantic Tension: A synthesis Two Views About Aid and the MDGs
“[T]he key question is how we can meet in practice the ambitious Millennium Development Goals. If every child is to have primary education, we will need $10 billion more a year. If infant mortality is to be cut by two thirds and maternal mortality by three-quarters, we will need at least $15 billion extra a year. If we are to halve poverty by 2015 we will need an additional $20 billion or more each year.”
-Gordon Brown, 2003
“Aid is just one of many important inputs to development, and the amount of aid that will be needed to meet the MDGs will depend critically on the quantity and quality of the supply of these other inputs. Indeed, the argument for targeting good performers grows out of the recognition that aid is most effective when coupled with good governance, and sound policy. ... Moving forward, we will need to present increased development assistance as a clear means towards an end rather than as an end in itself. This will require us (first) to define clear objectives for development funding and (then) to identify demonstrable results associated with those objectives.”
-John Taylor, 2005
Trans-Atlantic Tension: A synthesis
There is a resource constraint in poor recipient countries (per Gordon Brown)
Health expenditureper capita
(current US$)2002
Primary education expenditure per capita
(current US$)2002
Bangladesh 11 34Burundi 3 12Cambodia 32 18Eritrea 8 19Nepal 12 29Niger 7 26Sierra Leone 6 21Source: WDI (2005) and author's calculations.
Trans-Atlantic Tension: A synthesis
But external resources should be linked to results (per John Taylor)And should create, not undermine government accountability to citizens
Trans-Atlantic Tension: A synthesis
What the “cost” studies do really tell us:
The necessary “additional resources” are feasible: well inside what donors have already pledged
And, if all other conditions were adequate (institutions, incentives), additional resources would make a difference
The challenge: ensuring that new resources complement rather than substitute for incentives and institutions
Conclusion
Treat the MDGs seriously but not literally